Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Answered
Hello Everyone, When Performing Hyperparameter Optimization, Is There A Way To Set An

Hello everyone,
When performing hyperparameter optimization, is there a way to set an objective_metric as the value logged using logger.report_single_value method?
This values is reported under the title 'Summary' in the Scalars view, but it seems like the optimizer doesn't find it?

  
  
Posted 9 months ago
Votes Newest

Answers 3


@<1523701070390366208:profile|CostlyOstrich36> am I doing anything wrong here?

  
  
Posted 9 months ago

Hi @<1523703961872240640:profile|CrookedWalrus33> , this should be supported. How did you configure HPO?

  
  
Posted 9 months ago

@<1523701070390366208:profile|CostlyOstrich36>

from clearml import Task
from clearml.automation import HyperParameterOptimizer, UniformIntegerParameterRange, DiscreteParameterRange


task = Task.init(
       project_name="examples",
       task_name="HP optimizer",
       task_type=Task.TaskTypes.optimizer,
       reuse_last_task_id=False,
   )

task.execute_remotely(queue_name="services")

an_optimizer = HyperParameterOptimizer(
    base_task_id="c7618e30ff5c4955b4942971b410f72d",
    hyper_parameters=[
            DiscreteParameterRange("General/checkpoint", values=['5a6bea85858241f79a08ef23556e9e37',
                                                                 'f64fb9d419d14f11b74d17ebd2c38fd9']),
            DiscreteParameterRange("General/dataset_id", values=['0c5af65c496d45e591e3c475591764dd']),
            UniformIntegerParameterRange("General/seed", min_value=1, max_value=100),
        ],
    objective_metric_title="Summary",
    objective_metric_series="train_accuracy",
    objective_metric_sign="max",
    execution_queue="rtx3090",

    total_max_jobs=3,
    save_top_k_tasks_only=2,
    spawn_project="examples/hpo1",
    max_number_of_concurrent_tasks=2
)

an_optimizer.start()
an_optimizer.wait()
an_optimizer.stop()
  
  
Posted 9 months ago
630 Views
3 Answers
9 months ago
9 months ago
Tags